derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
136 lines
4.6 KiB
LLVM
136 lines
4.6 KiB
LLVM
; RUN: llc -mtriple=amdgcn -mcpu=tahiti -stop-after=instruction-select < %s | FileCheck -check-prefixes=GCN,GFX6 %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=fiji -stop-after=instruction-select < %s | FileCheck -check-prefixes=GCN,GFX8PLUS %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -stop-after=instruction-select < %s | FileCheck -check-prefixes=GCN,GFX8PLUS %s
|
|
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -amdgpu-enable-vopd=0 -stop-after=instruction-select < %s | FileCheck -check-prefixes=GCN,GFX8PLUS %s
|
|
|
|
; GCN-LABEL: name: s_shl_i32
|
|
; GCN: S_LSHL_B32
|
|
define amdgpu_kernel void @s_shl_i32(ptr addrspace(1) %out, i32 %lhs, i32 %rhs) {
|
|
%result = shl i32 %lhs, %rhs
|
|
store i32 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: v_shl_i32
|
|
; GFX6: V_LSHL_B32_e32
|
|
; GFX8PLUS: V_LSHLREV_B32_e32
|
|
define amdgpu_kernel void @v_shl_i32(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%b_ptr = getelementptr i32, ptr addrspace(1) %in, i32 %tid
|
|
%a = load i32, ptr addrspace(1) %in
|
|
%b = load i32, ptr addrspace(1) %b_ptr
|
|
%result = shl i32 %a, %b
|
|
store i32 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: s_lshr_i32
|
|
; GCN: S_LSHR_B32
|
|
define amdgpu_kernel void @s_lshr_i32(ptr addrspace(1) %out, i32 %lhs, i32 %rhs) {
|
|
%result = lshr i32 %lhs, %rhs
|
|
store i32 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: v_lshr_i32
|
|
; GFX6: V_LSHR_B32_e32
|
|
; GFX8PLUS: V_LSHRREV_B32_e64
|
|
define amdgpu_kernel void @v_lshr_i32(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%b_ptr = getelementptr i32, ptr addrspace(1) %in, i32 %tid
|
|
%a = load i32, ptr addrspace(1) %in
|
|
%b = load i32, ptr addrspace(1) %b_ptr
|
|
%result = lshr i32 %a, %b
|
|
store i32 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: s_ashr_i32
|
|
; GCN: S_ASHR_I32
|
|
define amdgpu_kernel void @s_ashr_i32(ptr addrspace(1) %out, i32 %lhs, i32 %rhs) #0 {
|
|
%result = ashr i32 %lhs, %rhs
|
|
store i32 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: v_ashr_i32
|
|
; GFX6: V_ASHR_I32_e32
|
|
; GFX8PLUS: V_ASHRREV_I32_e64
|
|
define amdgpu_kernel void @v_ashr_i32(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%b_ptr = getelementptr i32, ptr addrspace(1) %in, i32 %tid
|
|
%a = load i32, ptr addrspace(1) %in
|
|
%b = load i32, ptr addrspace(1) %b_ptr
|
|
%result = ashr i32 %a, %b
|
|
store i32 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: s_shl_i64
|
|
; GCN: S_LSHL_B64
|
|
define amdgpu_kernel void @s_shl_i64(ptr addrspace(1) %out, i64 %lhs, i64 %rhs) {
|
|
%result = shl i64 %lhs, %rhs
|
|
store i64 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: v_shl_i64
|
|
; GFX6: V_LSHL_B64
|
|
; GFX8: V_LSHLREV_B64
|
|
define amdgpu_kernel void @v_shl_i64(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%idx = zext i32 %tid to i64
|
|
%b_ptr = getelementptr i64, ptr addrspace(1) %in, i64 %idx
|
|
%a = load i64, ptr addrspace(1) %in
|
|
%b = load i64, ptr addrspace(1) %b_ptr
|
|
%result = shl i64 %a, %b
|
|
store i64 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: s_lshr_i64
|
|
; GCN: S_LSHR_B64
|
|
define amdgpu_kernel void @s_lshr_i64(ptr addrspace(1) %out, i64 %lhs, i64 %rhs) {
|
|
%result = lshr i64 %lhs, %rhs
|
|
store i64 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: v_lshr_i64
|
|
; GFX6: V_LSHR_B64
|
|
; GFX8: V_LSHRREV_B64
|
|
define amdgpu_kernel void @v_lshr_i64(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%idx = zext i32 %tid to i64
|
|
%b_ptr = getelementptr i64, ptr addrspace(1) %in, i64 %idx
|
|
%a = load i64, ptr addrspace(1) %in
|
|
%b = load i64, ptr addrspace(1) %b_ptr
|
|
%result = lshr i64 %a, %b
|
|
store i64 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: s_ashr_i64
|
|
; GCN: S_ASHR_I64
|
|
define amdgpu_kernel void @s_ashr_i64(ptr addrspace(1) %out, i64 %lhs, i64 %rhs) {
|
|
%result = ashr i64 %lhs, %rhs
|
|
store i64 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
; GCN-LABEL: name: v_ashr_i64
|
|
; GFX6: V_ASHR_I64
|
|
; GFX8: V_ASHRREV_I64
|
|
define amdgpu_kernel void @v_ashr_i64(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%idx = zext i32 %tid to i64
|
|
%b_ptr = getelementptr i64, ptr addrspace(1) %in, i64 %idx
|
|
%a = load i64, ptr addrspace(1) %in
|
|
%b = load i64, ptr addrspace(1) %b_ptr
|
|
%result = ashr i64 %a, %b
|
|
store i64 %result, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.workitem.id.x()
|